Perspectives on Retrospective Review and Analysis

I. INTRO

Chairman Coble, Ranking Member Cohen, and members of the subcommittee, I am pleased to be able to speak to you today about retrospective review and analysis of federal regulations. I am Randall Lutter, visiting scholar at Resources for the Future. My testimony today is based partly on a report I recently authored for the Mercatus Center at George Mason University, which I submit for the record, and partly on my experience managing and evaluating federal regulatory programs to reduce risk. I have served in senior staff and executive positions at the federal Office of Management and Budget, the Council of Economic Advisers, and the Food and Drug Administration. My work on different regulatory matters appears in a variety of scholarly journals. My testimony represents exclusively my own views, and not necessarily those of any organization.

II. BACKGROUND

While the federal government offers substantial protections to Americans’ health, safety, environment, and financial security, the specifics of federal regulation deeply frustrate many Americans as they try to read prescription drug labels or mortgage disclosure forms, travel by air, or manage small businesses. Codified federal regulations today total more than 165,000 pages and have grown at an average rate of 2.8 percent per year since 1970.

Concern over federal regulations has led President Obama to issue three executive orders on federal regulatory policy since January 2011.[2] All three spell out policies on retrospective review—the reexamination of extant regulations to identify modifications, including either increases or decreases in stringency, and possible elimination as warranted. My remarks today focus on such review, and on the retrospective analysis that can inform such review.

Today I emphasize two specific questions; the appendixes to my testimony provide supporting information and details.

III. HOW WELL HAVE RECENT REGULATORY REVIEW EFFORTS WORKED?

As described in my report for Mercatus, President Clinton’s National Partnership for Reinventing Government, under the leadership of Vice President Gore, was successful at reducing the number of pages of regulations. The total number of pages in the Code of Federal Regulations fell by about 7,000 from 1995 to 1997, leaving it at about 14,800 pages below where it would have been in 1997 if instead it had followed long-term trends. A count of pages, however, is not equivalent to regulatory burden or a measure of people’s welfare. Moreover, in a later listing of accomplishments for its first five years, the National Partnership makes no mention of any specific reduction in regulatory burden. Were the economic effects of this simplification of regulatory policy nil, because the changes in rules were simply housekeeping steps, such as the elimination of unnecessary and obsolete regulations governing buggy whips and horse-drawn carriages? Or were there genuine efficiency gains as a result of the modification or elimination of regulations that were recognized as inefficient with the benefit of hindsight? In fact, it is quite unclear what aggregate economic effects this reinvention initiative had on regulated entities or the public, because there was no claim of aggregate effects, and I am unaware of independent estimates of such effects by analysts outside the federal government.

During the George W. Bush Administration, the Office of Management and Budget (OMB) conducted different retrospective regulatory review efforts. I focus on one started in 2004. In that year, OMB launched a major effort to review existing regulations of the manufacturing sector and reported on this effort in its 2008 report to Congress. Its effort began with 189 “nominations” that members of the public provided in response to a request in a 2004 draft report to Congress for suggestions for specific reforms to regulations, guidance documents, or paperwork requirements that would improve manufacturing regulations. OMB determined 76 of the 189 nominations to be priorities and found that 69 of the 76 reform items were complete as of January 2009.[3] The items listed in that draft report are quite diverse and include actions better characterized as administrative steps rather than modifications or eliminations of federal regulations.

For example, the OMB draft report mentions 16 federal reports, at least two of which concluded that no change in existing rules was appropriate. They include four guidance documents, which do not have the full force and effect of law. They also include actions such as an Environmental Protection Agency determination,[4] a response to a petition,[5] a revised reporting policy,[6] the development of an “internal issue paper,”[7] one action that appears to substantially precede the retrospective review process initiated in 2004,[8] and a legislative action regarding taxation.[9] At least two of these actions, e.g., the Environmental Protection Agency’s Spill Prevention Control and Countermeasures rule and its Hazardous Waste rules to encourage recycling, likely offered significant savings.[10] Since OMB did not offer any aggregate estimate of the benefits and costs of these regulatory changes, however, there is little basis for a general judgment. Notwithstanding the paltry evidence of economic effects, this process of retrospective review offers other benefits. In particular, it gives regulated entities, nonprofit organizations, and the general public an opportunity to request changes to extant federal regulations.

These two historical examples suggest that the accomplishments of retrospective regulatory review efforts to date have been modest. One reason for the modest results may be that agencies in charge of developing regulations to achieve statutory missions have difficulty improving their own regulatory programs without specific stimulus from outside the agency. While such efforts have been able to reduce the number of pages of extant regulations, and sometimes led to revisions to existing regulations that reduce excess burden, whether they do this in a manner that significantly lowers overall regulatory burden or improves efficiency is hard to say.

IV. ARE AGENCIES TAKING ACTIVE STEPS TO IMPROVE MEASUREMENT OF THE ACTUAL RESULTS OF FEDERAL REGULATIONS?

In January of 2011, President Obama, in Section 1 of E.O. 13563, stated eloquently that the regulatory system “must measure, and seek to improve, the actual results of regulatory requirements.”[11] Moreover, in Section 1 of E.O. 13610, on identifying and reducing regulatory burdens, the President reiterated this same phrase. Many economists and analysts would agree that improved measurement of the effects of federal regulation is important. Such measurement could reduce the uncertainty about the effects and the merits of regulatory programs. If coupled to improvements, it could foster confidence that the federal regulatory system is delivering valuable benefits at modest costs. Unfortunately, agencies appear to be taking only very limited steps to improve such measurement.

Key federal agencies are not incorporating efforts to measure actual results of regulations into their major rulemakings. I considered the Federal Register notices for all economically significant proposed, final, and interim final rules issued by four key federal agencies since E.O. 13563 was published, on January 21, 2011. I focused on the Environmental Protection Agency (EPA), the Food and Drug Administration (FDA), the National Highway Traffic Safety Administration (NHTSA), and the independent Securities and Exchange Commission (SEC). I include the SEC even though it is an independent agency because President Obama’s July 2011 E.O. 13579 contains the same language as Section 6 of E.O. 13563, except that it uses the word “should” instead of “shall.”[12] I find that none of these rules issued by any of these agencies—25 in total—mentioned either “must measure” or “actual results.”[13] The word “retrospective” came up only once in the 25 rules examined and this occurrence was not relevant to the ideas in E.O. 13563. This finding indicates that these rules have been drafted without identifiable efforts to measure actual results of regulations, as the president had directed.

This is not to say that the agencies are not implementing the president’s directive to develop plans for retrospective review. In fact, they have issued and even updated such plans. As described in my report for Mercatus, however, such plans, at least for the four key agencies that I focus on, mostly reflect business-as-usual management. There is little discernible new work on the retrospective analysis and measurement called for in the executive order.

It is worth noting that information regarding retrospective estimates of the benefits and costs of federal regula- tions is quite limited. In a 2011 report to Congress on the benefits and costs of federal regulation, the OMB uses the phrase “benefit-cost ratios” in a discussion of retrospective analysis, suggesting that it has information on both costs and benefits for a set of regulations.[14] In fact, a footnote in an earlier 2005 OMB report that also used ratios acknowledges that this is rarely the case, saying, “A ratio was used [in the OMB analysis] because in most cases benefits were not monetized and, in some cases, unit benefits were not projected for health or environmental improvements.”[15]

In my own review of these retrospective analyses of extant federal regulations, I found very few that provide sufficient information to evaluate whether benefits outweighed costs. The overwhelming majority of retrospective analyses reviewed by Winston Harrington, the OMB, and later work by David Simpson of EPA provide information only about costs, about a key but incomplete measure of benefits (such as fatalities but not nonfatal injuries), or about both costs and a poor proxy for benefits (such as emissions reductions or the number of acres treated by a pesticide). In reviewing those retrospective analyses, I identified just four regulations, all issued by the NHTSA, for which retrospective studies provided both information about costs and reasonably comparable measures for benefits, expressed either in terms of dollars or in adverse health outcomes avoided. For another five regulations issued by the Occupational Safety and Health Administration, there are retrospective studies of reduced fatalities attributable to regulations. Unfortunately, these retrospective studies provide no estimates of the nonfatal injuries avoided or of the costs of the control technologies. The rest of the retrospective studies provide insufficient information to make judgments about the regulations’ economic merit.

This observation suggests that a key focus should be on improving the availability of information about costs and benefits, or at least about effectiveness. One approach would be to use more regularly market-based regulatory approaches such as tradable permits. Such permits provide information about the marginal cost of controls through their market prices. A second approach would be to use studies of consumer comprehension, including through the random assignment of surveys of different design, to judge the effectiveness of mandatory information disclosure, an important strategy in both environmental and financial regulation. A more challenging longer- term approach would be to foster a culture of experimentation, so that regulatory agencies have more information about the likely costs, effectiveness and benefits of their actions before promulgating final rules, although such a goal may be attainable only in the relatively distant future.

V. RECOMMENDATIONS

I have two general suggestions for policy makers:

1. Promote impartiality in retrospective analysis and review. Refrain from asking for self-evaluations and self- review and instead seek review and analysis from independent third parties. Government officials, naturally, have difficulty being objective in the review or analysis of regulations that they or their colleagues earlier developed. 
2. Seriously promote data collection, access, and analysis, either by incorporating them into new regulations being issued or by other means. Use of market-based approaches can automatically generate information about the marginal costs of controls; well-functioning markets for sulfur dioxide emission controls, for example, provide current information about the incremental cost of reducing such emissions. Studies on consumer comprehen- sion of information being disclosed might reveal that alternative forms are much more effective at promoting comprehension. More generally, greater data collection, access, and analysis can foster improved understanding of regulations and contribute to reducing regulatory burdens and improving their effectiveness. 

Thank you for the opportunity to testify today. I look forward to your questions.

Footnotes

<sup>1. I gratefully acknowledge the very helpful assistance of Asa Skinner in preparing this testimony.
2. Executive Order no. 13563, Code of Federal Regulations, title 3, sec. 3,821 (January 18, 2011), http://www.reginfo.gov/public/jsp/Utilities/ EO_13563.pdf; Executive Order no. 13579, Code of Federal Regulations, title 3, sec. 41,587 (July 11, 2011), http://www.reginfo.gov/public/ jsp/Utilities/EO_13579.pdf; Executive Order no. 13610, Code of Federal Regulations.
3. U.S. OMB, “Appendix E” in 2008 Report to Congress on the Benefits and Costs of Federal Regulations and Unfunded Mandates on State, Local, and Tribal Entities (January 2009).
4. Item #36 in Appendix E to the 2008 report, see also Item #75, to streamline and simplify duty drawback at customs.
5. Item #51.
6. Item #68.
7. Item #116.
8. Item #145, published as a final rule in December, 2004.
9. Item #188.
10. See Item #42 (73 FR 64668) and Items #54–#58 (73 FR 74235) in OMB’s draft report.
11. Executive Order no. 13563.
12. Ibid.
13. The EPA issued eleven rules, the FDA issued two, the NHTSA issued three, and the SEC issued nine.
14. U.S. OMB, 2011 Report to Congress on the Benefits and Costs of Federal Regulations and Unfunded Mandates on State, Local, and Tribal Entities (June 2011), 64.
15. U.S. OMB, Validating Regulatory Analysis: 2005 Report to Congress on the Costs and Benefits of Federal Regulations and Unfunded Manda- tes on State, Local, and Tribal Entities (2005), 42.</sup>